Abstract The clinical relevance of preclinical radiobiological studies depends on their ability to replicate state-of-the-art human treatments, which have achieved high targeting accuracy and dose conformality based on technological breakthroughs including intensity modulated radiotherapy (IMRT) and image guided radiotherapy (IGRT). Despite recent developments in preclinical irradiators for providing IGRT and more accurate dose calculations, a critically important component in modern radiotherapy, IMRT, is still missing. Without IMRT, the animal studies are unable to represent the dosimetric quality of state of the art human treatments, particularly with current trend of employing more complex, concave or simultaneously boosted dose distributions. As a result, successful translation from animal studies and human trial is rare. Previous efforts to develop IMRT and better mimic state of the art human radiotherapy have been unsuccessful due to the difficulty of miniaturizing multileaf collimator (MLC) and providing a planning system that is intuitive to use by biologists. To overcome these challenges, we will develop a global convex optimizer for beam orientation selection and rectangular aperture optimization so high quality IMRT treatment plans can be automatically and efficiently created using only rectangular apertures. We then propose a novel small animal IMRT dose modulator enabled by a breakthrough in plan optimization to efficiently deliver uncompromised IMRT plans with higher achievable resolution than a theoretically miniaturized MLC. We will make the hardware design and software open source to the research community. To achieve the goal, we propose the following aims: Aim 1. Develop a preclinical treatment planning system that automatically selects beam orientations and optimizes intensity modulation using multiple rectangular apertures. Aim 2. Design, fabricate and validate a sparse orthogonal collimator (SOC) to deliver the optimized rectangular segments.